© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 123069
    Prev
    Next
    INDEX
    Explanations

    phrases indicating causation or results

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    preposition + which

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     EconPapers
    -0.59
    HomeAsUpEnabled
    -0.54
    BeginInit
    -0.51
    DebuggerStep
    -0.46
    +#+
    -0.45
    :✨
    -0.45
     AssemblyCompany
    -0.43
    Portail
    -0.43
    migrationBuilder
    -0.43
    发表于
    -0.42
    POSITIVE LOGITS
     which
    1.06
     whose
    0.88
    which
    0.81
    whose
    0.77
     whom
    0.74
     WHICH
    0.73
     której
    0.64
     którym
    0.64
     whereby
    0.62
     waarvan
    0.62
    Activations Density 0.053%

    No Known Activations